Visor 1.3.0

New

• Choose how long the glasses wait for an answer. Settings > Meta AI > The glasses wait for an answer offers 20, 30, 45 or 55 seconds. The glasses used to give up after 20 seconds and say "Open the Meta AI app", which cut off answers that needed a web search, a smart home action or a slow model. Visor now gives them 45 seconds unless you choose otherwise.
• The assistant reads the text you look at. When you ask about something in front of you, the glasses read its text from the photo, and Visor now gives that text to your model. A model without vision can now answer questions like "what does this say?".

Fixed

• A slow model no longer keeps the glasses hanging. When a model took too long, Visor kept waiting after the glasses had already stopped listening, so the answer never arrived. Visor now stops in time and the glasses say that it could not get an answer.

Full Changelog: v1.2.3...v1.3.0